Talk:Global Positioning System/Archive 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Wikipedia and GPS

Wouldn't it be useful to have GPS coordinates in all Wikipedia articles that are about locations? GPS links could be treated specially. For example, clicking on then could take the user to a map. --HelgeStenstrom 19:44, 9 Feb 2005 (UTC)

There's a project to do just that. Have you checked out any of the Rambot articles (say, Autaugaville, Alabama) lately? --Carnildo 20:05, 9 Feb 2005 (UTC)
The project is located at Wikipedia:WikiProject Geographical coordinates. Shawnc 02:19, 19 November 2005 (UTC)
I added a link to this in the See Also section of the article -- Michaelfavor 18:42, 11 February 2006 (UTC)

GPS Jamming

This might be a nice addition "In fact, we destroyed a GPS jammer with a GPS weapon," US Major General Victor Renuart told reporters at a briefing in Qatar Commking 21 October 2005

Which means the use of a GPS jammer is a superb idea! Electronics are very cheap to manufacture nowadays. GPS jammer plus ancient russian truck to run it around costs 1000x less then the command shelter or nuclear reactor or Ministry of Propaganda palace it is used to protect. It costs 50x less then the fighter-bomber and GPS-guided bomb America drops to destroy the jammer. So you can and you must deploy dozens of GPS-jammers and as long as one remains running the true asset is protected. Meanwhile USA will run out of budget dropping thousands of smart bombs on every jammer in sight. Also, one country could launch big GPS-jammer satellites in geostationary to black out its own territory and that would be difficult to destroy. —The preceding unsigned comment was added by 195.70.32.136 (talkcontribs) .

What do GPS Coordinates look like?!

Can somebody post, as examples, GPS coordinates for some popular landmarks?

Cklester 16:47, 14 November 2005 (UTC)

Well, that depends on what format you want them in. Here are some samples:
N12.3456° W78.9012° — Degrees only (D)
N12° 34.567' W78° 12.345' — Degrees and minutes (DM)
N12° 34' 56.789" W78° 12' 34.567"; Degress, minutes and seconds (DMS)
12T E 408858 N 4569945 — Universal trans mercator (UTM)
These do not represent the same locations, but these do show some of the coordinate formats that you could have from a GPS unit. Some GPS units will let you switch between these different formats. Geocaching.com uses DM, or UTM. Google Earth uses DMS. Other places use degrees only.
If you wanted actual coordinates for popular landmarks, you'd also have to specify what Earth model (datum) you want to use for the coordinates. WGS-84 is the most up-to-date, so it is the one used, but my Garmin can be set to use any of over a dozen others. Val42 06:05, 15 November 2005 (UTC)

GPS Accuracy

According to the 2001 Federal Radionavigation Systems report published by the United States Department of Transportation (DOT) and the United States Department of Defense (DOD) the Standard Positioning Service (SPS) is stated to have a global average predictable accuracy, 95% of the time, of 13 m horizontally and 22 m vertically. I'm curious as to where this article gets it's accuracy numbers from. I figure since DOT and DOD operate the system they would best know the accuracy of the system. I'd like to see some references cited for the accuracy numbers used in this article. Sure the SPS might be accurate to 5 meters, but how often, 1%, 10%, 50% of the time? Is this an average number or a best case number. See also 2001 SPS Performance standard (PDF) (2,101KB) --Dual Freq 00:47, 16 December 2005 (UTC)

In personal experience, with GPS receivers employing the current SirfStarIII chipset, an accuracy of better than 5 meters/15ft is the norm. Don't mind my thumb :) Kar98 00:30, 17 December 2005 (UTC)
I'd be very careful in believing what a GPS unit outputs for it's estimated position error (EPE). This is explained in a fairly decent manner here (disclaimer:I have no affiliation with this site). I can summarize by saying the EPE displayed by a GPS is pretty much just a guess based upon the dilution of precision of the satellites. Even the wikipedia article talks about error sources that are larger than 5 m. For example, this article states that ionospheric delay can account for errors greater than 10 m by itself, that's not counting the rest of the errors that an SPS single frequency user will have. How can the same article say GPS has an accuracy of 5 m in the introduction? This site (same disclaimer as above) goes into exhaustive detail on the issue of error. I guess the bottom line is, I think this article should provide accuracy numbers that have some sort of reference, not just what number a hand held GPS tells them most of the time. --Dual Freq 03:06, 17 December 2005 (UTC)
I quite agree, which is why I made sure to mention that I was talking from personal experience alone. Mark it down as Anecdotal_evidence :-) That's just me saying comparing the indicated position on the read-out with the actual position in the field, current consumer grade GPS receivers (tested a few, not just one) seem to have a consistent accuracy of better then 5 meters, without WAAS/EGNOS augmentation. Further anectodal evidence tells me that surveyor-grade units deliver accuracy of better than .3m, or 30 centimeters (after post-processing). Kar98 03:26, 17 December 2005 (UTC)
I agree it would be nice if the article gave a better reference for the accuracy claimed. I read parts of 2001 SPS Performance standard, and the question doesn't seem to have a simple answer that fits well in an introductory paragraph. I notice that the article is claiming accuracy as of 2005, and the SPS Performance Standard is from 2001. The performance standard also states that accuracy is improving as older satellites are replaced with newer ones, and even in 2001 the average SIS (signal in space) error was considerably less than 5 meters. The performance standard also specifically acknowledges that it does not include information about additional accuracy that can be attained through the use of dual-band (L1 and L2) receivers. And while the article does says that ionospheric delay is the largest source of pseudorange measurement errors, those errors can be corrected mathematically after the rough position of the receiver has been established. Also considering that many new GPS receivers include WAAS capabilities, I think the accuracy claimed in the introduction is probably a little optimistic, but plausable.
If you don't believe the 5 meter claim, I encourge you to change it to 13 or 22 or whatever you prefer, and site your source. If somebody else has a newer source that says it's gotten better since 2001, the accuracy of the article will just keep getting better, too. --Michaelfavor 17:33, 10 February 2006 (UTC)
I made a note of the question in the article by adding a {{Fact}} tag, and a mention of the 2001 performance standard. --Michaelfavor 18:23, 10 February 2006 (UTC)
I guess I'm not sure what part of GPS is improving as the article states. SPS users have seen significant improvments since selective availability was disabled, is the article talking about that improvement? Is the article saying that dilution of precision is lower now because of more satellites thus increasing accuracy? Is the article saying that GPS satellite vehicle clocks are more accurate now because of some improvement? Is it saying that the ionosphere (the largest source of error) is no longer affecting the signal as much as it used to? Or is the article saying, my GPS receiver made by brand X says a lower error number than the one by brand Y and that means GPS is now more accurate? What has changed since 2001 that invalidates the accuracy numbers published by the DOD? I'd hate for this article to mislead users into thinking single frequency, non-differential GPS is some super accurate navigational system. A second civil use frequency in the future could improve accuracy, but that hasn't been implemented yet. Right now the second frequency isn't usable without the decryption keys held by DOD. Differental methods such as WAAS/DGPS improve accuracy, but plain old single frequency, what most of people reading this article are probably using, hasn't changed too much since 2001. Newer GPS receivers can compare the phase difference between the L1 and L2 frequencies to actually measure the atmospheric effects on the signals and apply corrections. Is this talking about survey grade GPS or plain old handheld GPS that most people are exposed to? --Dual Freq 22:38, 10 February 2006 (UTC)
FAA WAAS Fact sheet states "Using WAAS, GPS signal accuracy is improved from 20 meters to approximately 1.5 - 2 meters in both the horizontal and vertical dimensions." Interesting that this article says 5 m. Of course the FAA is worried about worse case and probably talking about 95% of the time. All I'm saying is the 5 m number should be qualified by whomever put it there. Is it a 50% time number? Is it anecdotal? Is it based on EPE numbers from a handheld GPS? --Dual Freq
I have made several change to the paragraph in question. How do you like it now? --Michaelfavor 15:18, 11 February 2006 (UTC)
Looks better, I might be nitpicking a bit here, but the DGPS article talks mainly about the USCG version and similar non-US systems which are listed as 1-3 m accuracy [1]. The centimeter systems are probably survey grade post processed or RTK and not USCG DGPS, right? --Dual Freq 15:43, 11 February 2006 (UTC)


Can Triangulation be implemented to improve the accuracy of GPS receivers, while also increasing the number of receivers.

Correction for Military section

The article currently states the following:

Civilian GPS receivers are required to have limits on the velocities and altitudes at which they will report coordinates; this is to prevent them from being used to create improvised missiles

I believe this is an incorrect understanding, and offer the following instead:

Due to the potential for GPS receiver technology to be used in improvised weaponry, the US Government has classified civilian receivers for controlled export. This means that, in general, a US-based manufacturer can not export a receiver unless it has limits on the velocities and altitudes at which it will report position and speed information.

The same source can be cited; its just the arms control information. Davandron | Talk 19:19, 13 November 2006 (UTC)

Locative games

There should be a link to locative games. The current version only includes Geocaching which is undoubtedly the most prominent example of a locative game, but there exist many others as well, especially in research. --Dendrolimus 19:57, 9 August 2006 (UTC)


Misc

Request to add clarifications:

1. It is stated that NM takes 12.5 minutes to transmit. There needs to be an explanation of how a GPS receiver can determine a satellite's position in only a few seconds after being turned on.

2. Please describe what good the P-code could be if that transmission repeats only once a week. -- JM 7 September 2006

Next time add your comment to the bottom of the page. Typing GPS "Navigation message" into google might help you understand a bit better. The first link is http://www.colorado.edu/geography/gcraft/notes/gps/gps.html and it talks about the nav message and its contents. It contains satellite almanac information that your receiver uses to locate satellites more rapidly. It also contains clock data and ephemeris data for the satellite transmitting the data. It's important information, but doesn't change frequently enough to require instant updating, plus there is only so much bandwidth available for data. Position is not determined with just the nav message, that is explained in this article at Global_Positioning_System#Calculating_positions. Dual Freq 00:05, 9 September 2006 (UTC)
I do not agree that the Wiki article explains how the positions are calculated, except by using the NM signals. It states: "It then looks up the ephemeris data for each satellite, which was captured from the NM and stored in memory." The colorado.edu page you mentioned above is also vague on this point and, anyway, I believe the Wiki article should clarify this since it dances around the point. That is, how can a data message with a 12-minute period be useful to provide a quick position fix? And what possible use is the P-code with a weekly period? Those questions leapt off the page at me. I believe your Wiki article is deficient in these matters. JM 216.165.144.128 20:49, 9 September 2006 (UTC)
Pseudorange lines. There is only one spot in the world located at the 3 calculated distances from satellite 1, 2 and 3
OK, this is a public encyclopedia, feel free to create an account and make some clarifications to the article. The entire nav message is 12.5 seconds long. databits.gif on the U Colorado site shows a sample 12.5 minute nav message. 1 subframe is 300 bits taking 6 seconds. each subframe begins with the 30 bit TLM, telemetry word. TLM contains 24 bits of data representing the current GPS time that the message was transmitted. So, the satellite transmits its time every 6 seconds. Your receiver uses that time, along with the times of 3 other satellites to determine it's location. With the times known, you can calculate how far each satellite is from you. Once you know how far the satellites are, imagine a sphere surrounding each satellite by the range you calculated. There is only one spot in the world where those three spheres will come together. I'm sure a GPS expert will criticize this oversimplified explanation. The P-code is only changed weekly and is used for anti-spoofing to prevent someone from faking the signal. I doubt that GPS units require very frequent crypto loads, I can't imagine a seal team or other military unit would pull out of a combat situation so the could go home and reload the crypto, so the decryption keys in a PPS receiver probably last more than a week. Also, I imagine the administrative problems of loading crypto keys into every vehicle, tank, plane, ship and portable unit in the entire DOD would prevent it from being feasible to require a new key if it was a short interval. FAS copy of FM 44-85 says: "They are a group unique key (GUK) and a cryptokey weekly (CKW). The GUK is normally good for a year, while the CKW is good for seven days."[2] This article is not intended to be a physics text on GPS system design, but to provide a basic overview of GPS. Dual Freq 22:09, 9 September 2006 (UTC)
("The entire nav message is 12.5 seconds [sic] long" -- I assume that's a typo.) You know too much and I know too little, so you're probably reading more into my original comment than I intended. If I knew the correct information to add, I'd do so. But I'm no expert, just an electrical engineer who recognized an inconstancy in the technical information section.
Above, you offered new and useful information: that the time is transmitted repeatedly and often enough in the 12.5-minute message that the time will be known within a few seconds. And I understand that from that you can determine distance to the satellites. But, as the article itself states ("By comparing the two, position and range, the receiver can discover its own location."), we need to know time (or distance) AND satellites positions. Are you saying you don't need to know the positions of the satellites -- that by visualizing the spheres alone there's only one place the intersection can be? Surely this is not the case; or if it is, then the Wiki quote is incorrect.
With just the time/distance, all you can determine is your position relative to the satellites, not your position relative to the Earth. If I'm wrong, then the Wiki statements (that we need to know where the satellites are) is incorrect.
The Wiki article states that the NM is 12.5-minutes long. In my view, that level of detail calls out for a further explanation of how such a slow-moving message could be useful. It is, after all, a "Technical description" section. If you believe that the details I'm requesting are inappropriate for Wiki, then I assert that you shouldn't be be priming the pumps with such minutia as bit rates and the like. You can't have it both ways.
Perhaps the answer is to add one of the following:
  • "In order to facilitate quick acquisition, the time is repeated every six seconds and the position is repeated every 30 seconds [or whatever] during the entire 12.5-minute navigation message."
  • "In order to facilitate quick acquisition, the time and position information are repeated numerous times throughout the entire 12.5-minute navigation message."
If that's true, then I propose one of those be added to the article and I'd be satisfied. JM 216.165.144.128 02:44, 11 September 2006 (UTC)



The www.gpsnews.org link is questionable. The change at 11:39 on Mar 24, 2006 from 151.37.154.185 did nothing useful - just added this link at the top of "Other Information". Is this the kind of thing that should be removed? It doesn't seem to me to add anything useful, and in any case is only "news" related to GPS receivers, not the GPS itself... -- JohnRobert 12:53, 18 April 2006 (UTC)


Anyone have a Code vs. Carrier Phase description?

There is a good explanation of this at Code-Phase GPS vs. Carrier-Phase GPS -- Michaelfavor 16:45, 23 February 2006 (UTC)

Should the Geocaching section say that the pasttime is popular with "both children and adults"? Surely that makes it popular with everyone, and that phrase is therefore redundant?

Check my change. Is that good? If not, you can fix it- be bold. Happy editing.--Adam (talk) 16:38, 14 February 2006 (UTC)



Is it really necessary that the receivers have a clock with good short term accuracy? I thought that the signals from the satellites are such that by simply comparing them, the receiver can tell the time difference, without much of a clock. For instance, if all satellites send the sequence 12345678901234567890... synchronously, with one digit coming every millisecond, and the receiver sees two sequences which have an offset of 3, then he knows that the signal from the one satellite travelled 3 (or 13, or 23...) milliseconds longer than from the other, without any need for a clock.

But I don't know the implementation details. Do we have any experts here? AxelBoldt

Messages are typcally referenced against an on-board quartz crystal oscillator/clock, which provides a timebase that all the others are compared to. The signals are individually weak, fading, full of interference -- comparing them directly with one another is harder than receiving them all independently, and comparing. Single channel receivers (now mostly obsolete) also had no other way of doing it. The Anome

I see, thanks. AxelBoldt

But, to be sure, the "short term accuracy" needed is not like a cesium clock. The internal crystal clock that The Anome is talking about needs to keep from drifting more than, say, a few nanoseconds (or tens of nanoseconds) during the time needed to "take a fix". Each nanosecond of drift corresponds to a light-nanosecond (about 30 cm, or one foot) of position inaccuracy, so 100 ns of drift means 30 meters of error - more than most modern units allow, typically. (Somebody check my math; I may have dropped a decimal point somewhere.) Also, since crystal frequencies vary with temperature change, all modern receivers (so far as I know, but I don't claim to be an expert) keep track of their crystal's behavior at various temperatures (with an internal thermometer), even when the GPSr is turned off, so that it will "wake up" with a fairly good resistance to drift. - Rootbeer 2002-04-07

Hence the importance of the Allan variance -- and Quartz crystals are really quite good clocks over suitably short periods (they're not bad over longer periods, either...) But that's a subject for another article. The Anome



Some officials believe that jammers could be use to attract the precision-
guided munitions towards noncombatant infrastructure, other officials believe 
that the jammers are completely ineffective. 

That sounds like a very interesting debate. Do we have sources? I'd be interested in knowing more.

Best, --257.47b.9½.-19 00:12, 4 Apr 2004 (UTC)


Question about relativistic corrections of satellite time (shouldn't they be rather opposite?).

The article reads:

"The clocks on the satellites are also affected by both special, and general relativity, which causes them to run at a slightly faster rate than do clocks on the Earth's surface. This amounts to a discrepancy of around 38 microseconds per day, which is corrected by electronics on each satellite."

Why is it slightly faster and not slower rate, and why 38 us/day while, according to my back-of-envelope math, the SR effect for the fastest possible satellite (at 8 km/s) is slowing of time by about 30 us/day and GR effect is rather negligible (and subtracting from those 30 s/day). Is it only my math? And if so could anyone provide the right one? Jim 18:05, 2004 Jul 8 (UTC)

SR slows the clock by about 7 us/day. GR speeds it up by about 45 us/day. The article is correct. See http://www-astronomy.mps.ohio-state.edu/~pogge/Ast162/Unit5/gps.html Amcfreely 14:23, 18 October 2005 (UTC)


" The receiver computes the distance to each of the four satellites by the difference between local time and the time the satellite signals were sent (this distance is called a pseudorange ). It then decodes the satellites' locations from their radio signals and an internal database. The receiver should now be located at the intersection of four spheres, one around each satellite, with a radius equal to the time delay between the satellite and the receiver multiplied by the speed of the radio signals. However, since the receiver does not have a ultra-precise reference clock, The intersection point gives the precise location of the receiver. If elevation information is not required, only signals from three satellites are needed. "

This does not make too much sense to me: without a highly precise local time (i.e. an atomic clock), one does not know the time delay with respect to a satellite, but only the time differences between the messages from the satellites. This yields 3 hyperboloids of revolution, not 4 spheres. I've checked on various sites and they also talk of hyperboloids. David.Monniaux 08:12, 18 Jul 2004 (UTC)

If the receiver had a highly precise local time it would ony have to use 3 satellites to calculate a position in space as a intersection of three spheres. Actually it get 2 point where the spheres intersect but one of them is far out in space (3-6 earth radius). Using four sattelites the receiver can also calulate the fourth dimension time and get an accurate time. I think the talk about hyperboloids is an other way to look at it as the interferens pattern of the phase of the radio signal as the DECCA or Loran system did. And with 2 diffrent frequencies the receiver is able to make corrections between them.

User Biker from Swedish Wikipedia [3]

Uh, just to clarify a little. If you have a pair of satellites with a difference in transmission time between them, you have two foci and the difference of your distance between them. The set of points where the difference to the distance between two fixed points is constant is a hyperboloid in 3D. If you had precise local time, then you'd have the intersection of two spheres - which is NOT two points (one far in space). The interesection of two spheres is a circle. --Rich0 19:12, 20 Nov 2004 (UTC)



The discussion of the L2 frequency needs to be cleaned up. At one point, the article mentions its use with high-end receivers, but in other locations it sounds like the signal is available for military applications only. The article fails to discuss the multiple segments of the GPS system: control, space, and user. I shall try to make some changes over the next few days once I have a chance to review my notes. Richss 05:28, Sep 6, 2004 (UTC)


Regarding trilateration vs. triangulation: By definition, trilateration seems correct. Doing a google search, it seems that many knowledgable sources also use this term. However, much of the literature that I have read regarding GPS and how it works previously, uses the term triangulation. I would suggest that perhaps we keep trilateration, but mention this discrepency to alleviate some confusion. Richss 14:18, Sep 6, 2004 (UTC)


There is a mistake in this article. reference points, whose distance differences to the unknown position are known, give (semi-)hyperboloids. 3 points are not enough in general, since all hyperboloids are mirror-symmetric to the plane in which the points lie and therefore the solution set is mirror-symmetric to this plane too.

System reliability

The article that is the basis for this section, GPS users must plan for outages, is basically FUD in support of the competing Galileo system. There are currently at least 6 on-orbit spare SVs. As of late 2002, there were 14 IIR SVs ready for launch. Lockheed Martin is building another 7 IIR-Ms and the IIF block is under design. See http://www.agcc.gov.au/gnss/GNSS_Platforms.aspx, http://gps.faa.gov/FAQ/faq-gps.htm, http://www.navcen.uscg.gov/gps/modernization/default.htm, and http://tycho.usno.navy.mil/gpscurr.html. Amcfreely 02:20, 16 October 2005 (UTC)

In the light of the above, I would trim the existing contents of the System Reliability section which I added simply to state that the reliability has been questioned, then append the above figures in conjunction with information on the numbers of existing satellites and their ages. I'm not myself familiar enough with the program to make the change myself (it's reasonable to question reliability without deep knowledge, but unreasonable to defend it). With such a critically important system reliability does need very careful scrutiny and attention. In the case of the space shuttle program, reliability was undermined by assumptions and complacency, but the GPS figures above do seem reassuring.

I do think that a System Reliability section is needed for such a critical system, hopefully only to reassure. Of course it will have to be revised often, but Wikipedia can cope. Pol098 00:35, 17 October 2005 (UTC)

System reliability, continued

The GPS signal is more fragile than might be supposed. A network of at least 24 satellites is required for full coverage. Satellites cannot be repaired and have a limited life. As of 2005:

  • there are 28 satellites in orbit, of which 16 are already beyond their design life—the oldest have nearly reached twice their design life.
  • The failure rate is about two satellites per year.
  • The launch rate of new satellites is about two per year.

If the older satellites start to fail faster—which may well happen— and if the launch rate cannot be increased proportionately, full coverage could be lost. The article GPS users must plan for outages discusses this possibility from the perspective of a backer of the competing Galileo system. Fortunately, this is not a concern in practice for at least two reasons. First, the Air Force monitors the health of SVs closely and maintains at least one on-orbit spare in each orbital plane. Second, the Air Force could easily increase the launch rate with SVs and ELVs already on hand.

i cut and pasted this section from the main article to this talk page, in case anybody thinks the discussion is worthwhile, or wants to add references. as it is, the whole section is a little too speculative for my taste, and i dont think it belongs in the article. --Michaelfavor 20:15, 21 January 2006 (UTC)
My reply to the "reliability concern" is not speculative at all. It is documented that there are on-orbit spares, a number of spares on the ground, and easily the launch capacity to get them into orbit. For references, see the System reliability discussion up-page.Amcfreely 15:46, 23 January 2006 (UTC)
No offense was intended. My comment was about the section as a whole. I actually think your sentences add balance to the section, and improve it quite a bit. Here is my question: Is it worth-while to add references to support the sensible parts of this section, or is the whole section so "non-notable" that I could be completely eliminated from the article without objection?
In the up-page discussion, you seemed to be of the opinion that the article supporting the original assertions was "FUD". Another user says up-page it's an important question to consider in a system this critical, and maybe he is right. I could go either way, remove the whole section or add references as needed. The launch rate from 1990 through 1995 was about four satellites per year, and I suspect the only reason the rate is currently two per year is because the pattern is full, and the launch rate is limited to match the failure rate. The main article is not suffering from being too short. If we agree that this question is sufficiently non-notable, we can remove it and increase the signal to noise ratio. -- Michaelfavor 23:39, 23 January 2006 (UTC)
The particular article is FUD, though I do agree that reliability is an important concern in a system so widely used and relied on. Amcfreely 00:46, 25 January 2006 (UTC)
You should consider that rockets sometime fail or explode on launch, so having spare satellites on the ground does not mean they will arrive in orbit. A big nasty solar storm or a new supernova anywhere in our trusty old Milky Way could kill multiple GPS satellites in orbit. Also, GPS satellites are a valid military target, because they have secondary, but direct military functionality (watch strategic missile launches) so an advanced adversary could decide to shoot down one or more of them (say Red China in a few years' time in a Taiwan war). Or a console knight (privateer or north korean) could hack the system and bring the GPS down. You definitely need to consider what you would do in a GPS-less world.—The preceding unsigned comment was added by 195.70.32.136 (talkcontribs) .
Like any other modern military system, GPS satellites are hardened against radiation (natural and man-made). A solar storm will not kill them. A supernova close enough to destroy them would present much more serious problems for earth. GPS satellites orbit over 12,000 miles high, which is 80 times higher than the space shuttle flies. To my knowledge no nation (including the U.S.) has ever had an anti-satellite capability for that altitude. The GPS satellite constellation is highly redundant, not even including orbital spares. Just destroying a few doesn't greatly degrade the system. Therefore multiple repetitive intercepts would be needed, which would take considerable time. That would of course be an act of war inviting major retaliation, so (even if possible) it wouldn't be done lightly or because of a regional conflict. There is no rational basis for "hacking" the GPS system. If so an adversary could hack spy satellites and use them, or any other military system using encryption. Most current military and certified civilian GPS-aided navigation systems are designed to work without GPS. E.g, GPS-guided munitions generally have a strap-down inertial navigation system for fallback use if the GPS signal is lost. Aircraft using GPS don't rely on that as a sole method. None of this is new -- any navigation system can be lost at any time for many reasons. Although highly redundant, hardened and secure, this includes GPS. Therefore all mission-critical navigation systems and techniques (whether GPS or other) are already designed to sustain the loss. Joema 20:41, 25 May 2006 (UTC)

GPS / WAAS relocation

According to this Public domain press release a WAAS satellite is relocating and might affect GPS/WAAS usage on the east coast of the US.

The move of the AOR-W satellite to the west may have a significant impact on non-aviation users on the East Coast. In its new position, the AOR-W satellite will appear lower on the horizon to users situated at ground level. As a result, the satellite may be obstructed by mountains, buildings, trees, or other types of obstructions that sit to the west of the user.

Sounds like a temporary problem and they say it will be resolved by Fall of 2006. Is this something that merits mentioning in the WAAS or GPS articles? Even though it's temporary doesn't it say something about reliability? --Dual Freq 16:26, 28 January 2006 (UTC)

Talk page archive

Isn't it customary to archive talk pages rather than delete things from them? See also: Wikipedia:How to archive a talk page. --Dual Freq 03:14, 10 February 2006 (UTC)

I've been trying to follow the suggestions in Wikipedia:Refactoring talk pages, about editing talk pages in the context of signal-to-noise ratio. It says: "Think what the talk page is about and remove anything superfluous that would not help future editors". A couple of the talk-page sections I erased were unsigned comments, fairly non-sensical questions, almost gibberish. I also erased a couple of sections that were basically to-do list items that were already completed without any controversy. I didn't see how keeping them would be helpful to future editors. My intent was not to delete anything controversial. Please feel free to restore any parts you feel are relevant.
One of the wonders of Wikipedia is that the old content is never really deleted, because it is always available in the page history. This gives us the freedom to limit the noise. I believe one of Wikipedia's greatest strengths is the high signal-to-noise ratio, and erasing useless and outdated content is vital to maintaining that quality. --Michaelfavor 12:03, 10 February 2006 (UTC)
You may as well remove the comments I made above about WAAS and GPS Accuracy, neither seem to be useful and pretty much were ignored. I still think GPS accuracy is a bit overstated on the main page, but why should wikipedia use DOD/DOT published accuracy information when it can use more reliable anecdotal accuracy numbers. --Dual Freq 12:24, 10 February 2006 (UTC)
I think the your comment above about GPS accuracy is useful and should not be removed. I see it as a "to-do" item that has not been completed yet. If you feel strongly about it, why not simply edit the article yourself? --Michaelfavor 17:08, 10 February 2006 (UTC)

Bug in article history

The history page for this article shows the oldest revision as being on 2001-02-19 [4], yet that revision does not identify the article as a new article as some other contributions by the same user to, but is marked by the user as a minor revision, without even an edit summary, and the article seems to be fairly well developed. It seems unlikely to me that was actually the first edit of the article.

The "older version" link within that version of the article links to the 2002-08-02 [5] version of the article. Presumable, continued clicking on the "older version" link would eventually lead back to the 2001-02-19 version, and continue in an infinite loop. -- Michaelfavor 17:09, 13 February 2006 (UTC)

See [6]. The policy of keeping the page history did not exist in the beginning of Wikipedia's existence, so the page history of that time is lost. Pfalstad 17:54, 13 February 2006 (UTC)

GPS tracking

Discussion of GPS tracking has been moved to Talk:GPS tracking.

New L1, L2, and L5 signals/Galileo interoperability/E5ab

The information in, e.g., http://www.navcen.uscg.gov/cgsic/meetings/summaryrpts/45thMeeting/13%20CGSICAJVD2.ppt, http://gps.faa.gov/gpsbasics/GPSmodernization.htm, http://www.navcen.uscg.gov/gps/modernization/default.htm, http://www.gpsworld.com/gpsworld/article/articleDetail.jsp?id=12195&pageID=1&sk=&date=, http://www.cast.uark.edu/local/GEOG4593/readings/EngeAIAC03.pdf would make useful additions to the article. It's difficult to tell what the most recent status of the GPS signal modernization effort is. Is anyone more familiar with the details? Does the information belong on a separate page? Amcfreely 06:32, 26 April 2006 (UTC)

Sorry about deleting this note, I'm not sure what happened, but I'll try not to do it again. --Dual Freq 02:48, 27 April 2006 (UTC)
Found a good article; lots of details. Seems like civilan L2 and L5 should have their own sections, or even pages (gps modernization).

good info -> http://www.navcen.uscg.gov/gps/modernization/L5BriefB2001VanDierendonk.ppt

External links

The external link section of this article is way overboard for this topic. Wikipedia:External links has some guidance, but I think 90% of the external links should be dumped as External link spamming. I'm not sure how controversial this removal would be, so I'm asking for input here. We're not trying to make a list page here or boost someone's hitcount. Right now the GPS article is 43kb in size, and 10kb of that is external links, over 100 lines. --Dual Freq 18:00, 14 May 2006 (UTC)

Since there was no response I removed all of the links that did not directly add to the topic or were not cited in the refs. People can use a search engine to find mfrs of GPS equipment/software. This is not a place to sell GPS equipment. --Dual Freq 00:54, 16 May 2006 (UTC)
I agree with the deletion of the links, although I think if you're "asking for input" then 31 hours is a pretty short time to wait! I'm guilty of adding one of the software links, although not because I've got any stake in the software, just because I use it and saw the list of other GPS software in this page. Maybe a new page would be useful, just about the receivers (including hardware and software)? 193.134.254.145 12:19, 16 May 2006 (UTC)
Thanks for responding, but I didn't wait very long mainly because I decided it was a defensible position to delete most of the external links. I based my decision on the above cited pages as well as other articles that had similar problems. For example, the Sudoku article has a warning message and fairly aggressive patrolling of the external links section. I think it was so bad they had to block IP's and restrict the page for a while due to the massive linkspam on that page. They have the warning NB: External links are being reviewed on the Discussion page ("Talk:Sudoku") at present. Please join in that discussion before adding new links here. posted in the external links section on that page and only 2 external links.--Dual Freq 22:51, 16 May 2006 (UTC)
As for the link page suggestion, I don't think an encyclopedia's role is to provide a list of external links beyond references. Any list would be very dynamic and could never list enough items to be unbiased. Its better to leave the reviews and lists to search engines. --Dual Freq 22:57, 16 May 2006 (UTC)
Totally agree with you. I wasn't suggesting a page of external links, but a page specifically about the receivers - different types, manufacturers, software that you can use with them etc. Not just a list of websites but for example links to Garmin, Magellan or TomTom and their ilk, plus (preferably internal) links to software, although there don't seem to be too many pages already (apart from GpsDrive). 193.134.254.145 08:41, 17 May 2006 (UTC)
I added the Powerpoint presentation the other day (wasn't logged in, my mistake), which you removed. You're probably right; it doesn't do a whole lot of good without a presentator, but it belongs with the PDF document and illustrates some of the techniques of GPS. Perhaps I will modify the page that links to the document to point to the presentation also. The external links list has been cleaned up nicely. Kudos! --Stimpy 12:58, 1 July 2006 (UTC)

Satellites using GPS

GPS seems to work even for (other) satellites to track their motions, such as GRACE. It would be great if this was also discussed, as I can guess this application has limitations (very high speeds, satellite outside GPS coverage). --Abdull 10:10, 5 June 2006 (UTC)

I was thinking orbital usage was limited to spacecraft lower in altitude than GPS, However http://tco.gsfc.nasa.gov/ft-tech-GPS-NAVIGATOR.html says there are high Earth orbit (HEO), geostationary orbit, and other high altitude applications. --Dual Freq 02:12, 13 June 2006 (UTC)

Citation

For large sections of new material it is recommended to cite sources per Wikipedia:Verifiability and Wikipedia:Citing sources. --Dual Freq 22:47, 14 June 2006 (UTC)

See also:Wikipedia:What is a featured article? and An article based on the knowledge in your head, even if you are the world's leading expert on your obscure Internet subculture, doesn't make it good reference document. The text has to be verifiable, and we can't fact-check your personal experience. - User:Jengod/Some common objections to featured status and how to avoid them. If this article is ever to become featured it must be sourced. Thanks. --Dual Freq 00:12, 16 June 2006 (UTC)

Relativity changes

Some anons keep changing the page to add in some text [7] casting doubt on the relativity and the relativistic corrections used by GPS receivers. Seems POV and original research to me. I'm not going to keep reverting them by myself, because of the 3RR rule. Any comments from the anons responsible or anyone else? Pfalstad 20:39, 25 August 2006 (UTC)

Well, it doesn't look like they are willing to explain themselves. I don't think this information belongs in this article, I guess the see also might be OK, but this article is 51K and adding a bunch of POV/OR and a link to Amazon.com is not going to help clarify things. I wonder if semi-protection is appropriate in this situation. Dual Freq 22:19, 25 August 2006 (UTC)
See Talk:Hafele-Keating experiment. I told him/her I agreed that the importance of the relativistic corrections may be exaggerated, but he/she isn't being reasonable. We should either semi-protect or simply keep reverting until the offender finds something better to do. —Keenan Pepper 22:52, 25 August 2006 (UTC)
This is "him/her". My main objection was that GPS was presented here (and in Hafele-Keating experiment) as a proof of relativity theory. This is wrong. The references provided such as the Leick's book are more than enough to any expert in the field. I am so sorry if what's written in that book bothers Star Track fans like the 19-year old Mr. Pepper, or even contradicts what his professor of physics might have told him. But he must learn that there are biased and opposing views amongst scientists just like with everyone else. If the cited book says: "relativistic corrections cancel out in relative (mm) positioning", it is quite obvious that such corrections were not needed in the first place, hence are not critical. Add a pink elephant, or a Bengal tiger if you like, but if in the next step you have to subtract it then it means that you did not need such a complication to begin with. Why is this so hard to understand for this teenager?? —The preceding unsigned comment was added by 200.253.18.10 (talkcontribs) 15:59, 26 August 2006 (UTC)
All you have is that one misquote from Leick's book. The rest is your original research/POV. In Leick's book, that "(mm)" is not there; that was added by you. Relative positioning does not mean millimeter-accuracy positioning. Rather, according to Leick, it refers to "relative positions between co-observing stations", as opposed to point positioning, which gives the position of a single station. Since the book is about surveying, it's not surprising that relative positioning is his main focus, which is why he doesn't care much about relativistic corrections. But he is not denying that relativistic corrections are important in point positioning: "The implications of relativity for GPS observables have widely been addressed in the literature... In this chapter, remarks on relativity are limited to section 5.3.1, where a clock correction term is given to account for satellite orbital eccentricity. The correction and the adjustment to the fundamental frequency... are the only references to the applicability of relativity to GPS in this book." (Also he says "most" of the corrections are negligible, not "all".) Pfalstad 17:14, 26 August 2006 (UTC)
Why my research or point of view? Just read the references I posted! And why do you call the quote from Leick's book a "misquote"? The "(mm)" emphasises the fact that if GPS relativistic corrections are not important for achieving mm accuracy, then of course they are NOT critical for the GPS system as well. Relative positioning certainly does include mm positioning (actually relative positioning is the ONLY way to achieve mm accuracy), so clearly there is no misquote -- just an emphasis. Back to the meaning of "relative": you can add a pink elephant or a Bengal tiger if you like, but if you immediately remove them why did you need those beasts in the first place? Relativistic corrections are an unnecessary complication and accurate positioning and navigation work just fine without them. —The preceding unsigned comment was added by 84.253.151.77 (talkcontribs) 18:52, 26 August 2006 (UTC)
References? You posted one, and I read it. GPS relativistic corrections are critical for determining one's position using GPS. What Leick is saying is that they are not critical for determining the relative position of one point compared to another point. So, you might be able to find the distance between two points to mm accuracy, but you cannot find your absolute position to mm accuracy, unless you apply the relativistic corrections. Before selective availability was removed, people used differential GPS to measure distances with high accuracy. The selective availability error canceled out when comparing two stations. But that doesn't mean that SA error was negligible. Pfalstad 20:39, 26 August 2006 (UTC)
This what you just wrote is not even wrong. Absolute position to mm accuracy?? Why in the world would anyone go to trouble of inventing and using (very expensive) relative GPS, if a thing like "mm absolute positioning" existed in the first place? The best a single handheld GPS unit can give you today (Selective Availability removed) is tens of metres. The best a single surveying GPS (GIS) unit can give you today is decimetres-to-metres (unless you have time to spend over a point -- for cm accuracy -- but even in that case it has nothing to do with relativity but with fresh ephemeredes broadcast from the base unit and containing latest atmosphere-corrected position differences). Your relativistic corrections are alleged to be at the centimetre-to-decimetre level. But not even mm positioning requires any such corrections, because whatever you add you also subtract during mm (relative!) positioning (see Leick). And I don't mind repeating this to you: relative positioning is THE ONLY way for obtaining mm accuracy using GPS! Finally, what in the world does SA have to do with relative positioning? Surveyors had been obtaining mm accuracy before the SA was removed, using differential (relative) GPS. But who misinformed you that they stopped using relative (differential) positioning after the SA was removed?? How dare you even discuss a topic (let alone delete entire posts) before learning the basics!? —The preceding unsigned comment was added by 84.253.151.77 (talkcontribs) 21:20, 26 August 2006 (UTC)
The error introduced if the clocks in the satellites were not corrected for SR/GR amounts to a drift of 11 km/day . Observe also that this correction for SR/GR is done in the satellites: no additional correction for these effects is necessary in the gps units themselves. —The preceding unsigned comment was added by Mossig (talkcontribs) 17:21, 4 October 2006 (UTC)

Ok, we agree on that then.. The issue is your changes. Putting time dilation in quotes? "According to Einstein's theory of relativity�?" "claimed by relativity physicists"? I hate to break it to you, but only cranks deny relativity at this point. I agree that relativistic corrections are not necessary when differential GPS is used; if you could add that to the article without adding the crank stuff about denying relativity, I would not revert it.. But relativistic corrections are necessary when not using differential (hand-held receivers for example) and are, in fact, used in commercial receivers (see Neil Ashby article). Pfalstad 22:08, 26 August 2006 (UTC)

"Cranks"? "Break it to me"? What the heck are you talking about, man? I don't remember agreeing on anything with you. First of, do you know what the word THEORY means? And that a theory is not to be taken seriously unless it was proven beyond any doubt AND earned consensus? Einstein's theory most certainly has not achieved either, except in the minds of people who do relativity for living(!), and an army of Startrackies who get high and fantasize about "wormholes" (tell me about cranks!). Get into that head of yours that there are hundreds of thousands of theories out there, but so what? A theory itself won't get you Nobel, just like the Einstein's theory didn't earn one for him. THIS JUST IN: Have you heard about NASA discovering dark matter, a few days ago? Did you not know that Einstein's theory of relativity (Yes, I do keep putting "Einstein's", but I do it out of respect for other people's theories of relativity; what, you never heard of those?) couldn't account for 95% of the mass of the Universe? And it certainly cannot explain (meaning: it does not predict at all) the existence of dark matter or dark energy. I don't know who you are (Wiki's anonymity is a great thing indeed) but it seems like you are the one who needs some serious brushing up here, and I don't mean just GPS. You should be reading Wiki rather than editing it; If I were you, I would start with Sir Isaac Newton. And stop locking pages that carry knowledge that frightens you. 11th century is long gone. —The preceding unsigned comment was added by 84.253.151.77 (talkcontribs) .
I just came to this talk page after reading the jarring and bizarre section on relativity on the GPS page. It goes from an informative and well written piece to a ranting piece of propaganda against the experimentally demonstrated fact that is relativity. The section needs severe cleaning. If no one else does, I shall do so myself in a few days. For the record, relativity is a theory which explains the way mass, space, and time interact. It is not concerned with explaining dark matter in the slightest (though it may turn out to be related to dark energy). This is no more a problem than a book about making cookies being unable to provide directions for cooking a steak. They are both true, but are about totally different subjects. Vonspringer 03:01, 12 September 2006 (UTC)
Haha, I'm glad I'm not the only one getting ranted at. BTW, why do you keep misspelling Star Trek? Is it a joke? Please explain. —Keenan Pepper 00:05, 27 August 2006 (UTC)
What's "Haha", some coded signal in trackies' sign language?? (Don't bother.) —The preceding unsigned comment was added by 200.253.18.10 (talkcontribs) .

Time dilation

Maybe the article about time dilation can be shorter and better.

"Because of their movement with respect to the Earth's reference frame the clocks on the satellites are affected by both special and general relativity. Observing from the Earth's reference frame, satellite clocks are perceived as running at a slightly faster rate than clocks on the Earth's surface. This amounts to a discrepancy of around 38 microseconds per day, when observed from the Earth. To account for this, the frequency standard on-board the satellites runs slightly slower than its desired speed on Earth, at 10.22999999543 MHz instead of 10.23 MHz—a difference of 0.00457 Hz.[13]

Neil Ashby presented an account of how these relativistic corrections are applied, and their orders of magnitude (more than 10 km a day), in Physics Today (May 2002).[14]"

Pukkie 12:03, 31 August 2006 (UTC)

Neil Ashby has agenda he is trying to sell. The consensus amongst GPS community is that GR and SR may add so-called relativistic corrections to the Newtonian mechanics as applied in the GPS system. For more, read the Alfred Leick's book (see GPS page for the ref.) to see why relativistic corrections are unnecessary even for the most accurate (mm level) GPS positioning. —The preceding unsigned comment was added by Uknewthat (talkcontribs) .
As it stands that section is probably WP:OR as it combines Neil Ashby and Alfred Leick in an apparently original way to "advance a position". In fact they may not even be discussing the same point (the new techinique of "relative positioning" may indeed not need relativity theory, but that is not quite what that section advances).
Thus, it should be shown that Leick there counters Ashby- as this article suggests - with the claim that GPS did not verify some of the predictions of relativity theory. Otherwise the text needs to be revised. Harald88 11:23, 16 September 2006 (UTC)
According to Keenan Pepper (Talk on Hafele Keeling experiment), on page 68 [Alfred Leick] says:
Relativistic effects are important in GPS surveying, but fortunately can be accurately computed. The atomic frequency standards in the satellites are affected by both special relativity (the satellite's velocity) and general relativity (the difference in the gravitational potential at the satellite's position relative to the potential at the earth's surface.
I conclude that the position that he advanced does not correspond with what this article now suggests. Harald88 12:52, 16 September 2006 (UTC)
No. Keenan is an Einstein's teenage groupie who takes quotes by researchers out of context and uses them to impose his own (a teenager's, mind you) view of what complex notions mean. Leick clearly stated in the same book (see the GPS section) that relativity corrections are not needed to achieve even the mm accuracy with GPS. —The preceding unsigned comment was added by Uknewthat (talkcontribs) .
In reality, the word even is missing... Harald88 00:22, 17 September 2006 (UTC)
And? — Preceding unsigned comment added by Uknewthat (talkcontribs) 20:11, 18 September 2006
I now added the lacking citation which results in a completely different account (and in which "even" doesn't fit). Harald88 21:27, 19 September 2006 (UTC)
No "completely different account", of course. Just your mumbo-jumbo. If A>B and B>C then sure as hell A>C. So that 'even' and 'mm' make a perfect fit in this case.— Preceding unsigned comment added by Uknewthat (talkcontribs) 02:37, 29 September 2006


I am starting an RFC on User:Uknewthat: Wikipedia:Requests for comment/Uknewthat. If you have anything to say about this user, please do it there. —Keenan Pepper 19:38, 16 September 2006 (UTC)

This person(s) now goes 'inquisition style'. A note to the serious readers/users of Wiki: I do not intend to discuss anything with an 18 year old Einstein's groupie, let alone sophisticated science. Have your milk Keenan, and then go to bed. You'll do the world a big favour. —The preceding unsigned comment was added by Uknewthat (talkcontribs) .
Uknewthat, I think there may be a bit of misunderstanding here. Perhaps you'd like to discuss your specific concerns about this article with me? I'm both an amateur GPS enthusiast and someone who has spent several semesters studying relativity under scientists who do both theory and experiment with relativity every day. It would be my pleasure to try to clear up some of the confusion you may have about the subject. As it stands, the section needs a great deal of work because of some assertions you have added which seem to be inaccurate. If you're interested, let me know and I'll post my email. Vonspringer 08:32, 17 September 2006 (UTC)
Confusion? You took a few semesters? Is that a joke? You can study relativity for eons if you like, but it will still be just a theory. So clearly, you are not as qualified as you imagine you are. "As it stands"? So, you are a judge somehow. Let me ask you this: are you with the same group as the persons under nick "Keenan"? Because I know there is a group out there that has been making concerted efforts to sell Einstein's nonsense to the world for ages now, but fortunately had succeeded only partially.— Preceding unsigned comment added by Uknewthat (talkcontribs) 17:12, 17 September 2006


Well, Newton's gravity is a theory as well. But it seems to be a true theory: if you drop the apple it will fall. Surely you do not object to that theory on the basis that it is in fact a theory? However, Einstein noticed a few places where Newton's gravity produced predictions that did not coincide with observations (perihelion precession of Mercury, failure of the Michelson-Morley experiment, invariance of c under Maxwell's theory of light, etc.). So he came up with the idea of relativity, which has so far described experimental observations of space and time quite well. You apparantly disagree. I am interested to know why, and am quite willing to listen to your perspective and share my own in a good-faith email exchange if you're willing. Just let me know. Vonspringer 17:12, 17 September 2006 (UTC)
Thank you for pointing out that Newton's theory is a theory as well. The fact that Einstein's relativity SOMETIMES agrees with observations to a higher degree of accuracy than other theories does not entitle the proponents of the Einstein's theory to lose quotation marks when selling produce like "time dilation", where "time" has no absolute definition to begin with. You cannot subject to relativity something that has no meaning at all. All you can do with it is abstract it, say "Do you have time?” "Time's up!” or represent it in RELATIVE terms as, say, particle orbit period. But since there is NO natural period in the entire Universe that is constant, why should Einstein's measure of "time" be allowed to be an exception i.e. a measure of anything, for heaven’s sake? Hence, you cannot make relative something that is ALREADY relative. Only ABSOLUTE concepts can be made (or be observed as) relative. By the way, it surprises me how many Einstein's groupies do not realise that not even Mercury perihelion prediction is entirely explained by relativity. Oh, I forgot, where did 90%+ of the mass of the Universe go? (Don't bother to explain.) Oops... — Preceding unsigned comment added by Uknewthat (talkcontribs) 20:13, 18 September 2006
Talk pages are not really the place for extensive debate about a subject, but rather what should be in the article. I'm perfectly willing to have an open-minded discussion about relativity with you (either by email or on my talk page). But here we just need to work out what should be in the article. The section you have edited contains statements that are not substantiated by any credible source. In fact, every professional scientist working on GPS or relativity can demonstrate that relativistic corrections are required for any level of accuracy over acceptable timescales. This has been extensively sourced. If you wish for the article to contradict this, you need to cite a professional scientist or GPS engineer who agrees with you. That's just how Wikipedia works. Also, please see the articles second and International Atomic Time for good discussions on how "time" is perfectly well defined, and how general relativity is required to understand why identical atomic clocks run at slightly different rates at different altitudes due to general relativity.Vonspringer 16:49, 19 September 2006 (UTC)
There is no such a thing as "identical atomic clock". In fact, there is no such a thing as "identical" in the material Universe at all, only in imagined ones. And that's precisely what Einstein's realm of relativity is -- an imaginary place filled with fairytales on worm/black/grey/etc. holes, "curved" "space-time", with Time and Length as main "living" characters, something like Mickey and Pluto in uncle Walt's world... Thanks for the offer, but I don't take lessons from youngsters. Uknewthat
Well, it's certainly up to you if you don't want to learn from a young upstart like me. But since it's a Wikipedia policy requirement to cite sources and include only verifiable information, I am going to rewrite the section appropriately, build consensus, and make the edit. If you object, you will need to provide credible sources and verifiable experiments with citations to support your position. Should you not be able to do this, under Wikipedia policy the article must take the consensus view, which is that relativity is real, verifiable, and required for the successful operation of GPS. Vonspringer 17:47, 20 September 2006 (UTC)
What is it, no more "arguments"?? I am disappointed -- not. Regarding the edits you are talking about: did someone beat you in doing just that? Because the way the article reads right now is quite satisfactory.— Preceding unsigned comment added by Uknewthat (talkcontribs) 18:59, 20 September 2006

POV?

I don't understand why this revert was done: [8]. I don't know if it's accurate, but it doesn't seem like POV to me. Pfalstad 02:35, 27 September 2006 (UTC)

2 of the edits seemed overly POV trying to describe GPS as a device used to aid in the targeting of nuclear weapons. I'm not sure that any nuclear weapons use GPS. TLAM-N is launched from ships with highly accurate INS systems, no need for GPS. TLAM-N used TERCOM, DSMAC and other tech, I don't think the nuke version used GPS. SLBM targeting via GPS? Subs have high accuracy INS as well, they are underwater and it is highly unlikely they would want to surface prior to attack and risk detection and possible destruction prior to launch. Its addition, to me, seems like a POV attempt to connect GPS to nuclear weapons and somehow condemn it. I've seen tis POV here before only last time it was pushing the idea that GPS made Trident more accurate so it could be used a a first strike weapon. The topo tick mark thing seemed unsupportable as well, what size topo are we talking about? 4 satellites is all you need for a 3d position, right? How is that position any less approximate than a 3d fix with 12 satellites, less accurate sure, but all fixes could be considered approximate depending on the accuracy needed. --Dual Freq 11:25, 27 September 2006 (UTC)
Expanding on what DualFreq said, SLBMs and associated targeting systems were designed long before GPS. That's why they use INS and some Trident missiles use steller updates (missile takes a star sighting while in space to correct for drift). However some sources state the Trident II D-5 steller navigation system has been revised to also incorportate GPS updates, thereby further increasing accuracy [9]. Other sources imply the GPS capability was only a test: [10]
Re 3D (vertical accuracy), in general altitude is less accurate than lat/long, at least for normal civilian use. I don't know about military P-code, which is more accurate than civilian GPS signals. Also GPS enhancements such as WAAS greatly improves vertical accuracy. Joema 11:48, 28 September 2006 (UTC)
Guidance systems are under constant development. A large number of conventional weapons use INS-stabilized GPS solutions. As to whether or not nuclear systems use the same system, I do not know. I do know, however, that INS-stabilized GPS-updated systems have been around for a good fifteen years, so it's entirecly conceivable that some nuclear systems may be INS-stabilized GPS updated systems. Using a purely GPS system is ludicrous, as GPS occasionally produces brief navigation errors. Imagine if such an error occurred during the descent phase of a MIRV. That's a lot of warheads off target. That's why both the GPS and INS signals are fed into a mathematical algorithm known as a Kalman Filter, which heavily weights consistent signals, ignores spurious ones, and accounts for variations common to all Earth navigation systems. The result is a sum that's significantly more accurate than the individual inputs of its parts. All modern GPS-aided INS systems aboard military aircraft, ship, and submarine, use variations of the Kalman Filter.
With respect as to whether this page should be unlocked or not, the answer is simple: Cite credible, verifiable sources, and the content should be allowed. Mugaliens 16:15, 6 October 2006 (UTC)

Page protection

I've protected because of Uknewthat's slow revert war. Let me know when it's sorted out and you're ready to start editing again, or leave a request on WP:RfPP. Cheers, SlimVirgin (talk) 01:26, 5 October 2006 (UTC)

Why did you revert the last change? That seemed like a good compromise -- leave the stuff what this chap writes about his grandfather's experiment, but ADD what I put about the GPS not being a definitive proof of GR/SR theories. The same applies to so-called "time dilation". What is the problem with that? Why revert that? Why lock without including what the other side has to say? —The preceding unsigned comment was added by Uknewthat (talkcontribs) 12:56, 5 October 2006 (UTC)
From WP:NPOV#Undue_weight: If a viewpoint is held by an extremely small (or vastly limited) minority, it doesn't belong in Wikipedia (except perhaps in some ancillary article) regardless of whether it's true or not; and regardless of whether you can prove it or not. Pfalstad 16:37, 5 October 2006 (UTC)
Is it possible to summarize, briefly and succinctly, exactly "what the other side has to say" regarding GPS without getting into namecalling and attacks? I'm sure that everybody involved in this revert cycle is willing to listen to any reasonable argument and review any evidence in support of any claims being made. Nothing seems to be advancing at this point. What's the conflict here, exactly? Justin 16:08, 5 October 2006 (UTC)
Maybe I should clarify. I'm hoping that Uknewthat will summarize his/her side of this conflict with material in the GPS article. I would like to see a brief description of what Uknewthat would like to change, and the justification behind it, without resorting to the sort of name-calling tactics displayed so far. Justin 21:35, 5 October 2006 (UTC)
Sure, here it is: 1a. Mr. Einstein Albert's theories of relativity are just theories, meaning they are not laws, so the public should not be mislead by persistently omitting quotation marks around the ever-ultramodern expressions such as "time dilation", "warped space", "worm hole", etc., or at least: 1b. Such theoretical terminology should be preceded and co-followed by disclosures like "According to Einstein's theory..." and similar, 2. There are other peoples' theories of relativity, such as the Poincaré's, which is also why one has to decide WHOSE theory of relativity one is talking about, and yes, even STATE it, 3. Wiki should not allow the use of widely disputed “arguments” (e.g., Hafele-Keating experiment; or GPS in which simple geometrical differencing renders any “relativistic corrections” unnecessary) as proofs of the Einstein's theories, precisely because those claims have been widely disputed (no doubt the Nobel committee would have given Mr. Einstein a prize for his theories had they been proven laws), 4. Acknowledge that in reality (leave magical worlds to Mr. Disney), Mr. Einstein Albert’s theories of relativity have not been proven as of yet, i.e., a century or so since they were first proposed, which decreases the chances for proving them at all. 5. Take those theories for what they are along with other (thousands upon thousands of) theories out there that make some correct predictions in some cases but then also make incorrect or no predictions in some other cases as well, 6. Learn to live with the fact that a patent clerk will remain a patent clerk no matter what the media and the rich tell you, 7. In case that you cannot live with all of the above in the real world simultaneously, then please do feel free to publish a scientific paper (in the real world, of course), in which you are going to define Time in absolute terms, where you will prove it is possible to change the size of the Time you previously described in absolute terms. An experimental observation of the quantity named Time should speed publishing of your paper tremendously.
Thanks, I think that should settle it for just about everybody who is interested in your perspective. Justin 03:56, 8 October 2006 (UTC)
You mean to say that there is an alternative to logics and truth? Please, do enlighten us!
Yes.. The threshold for inclusion in Wikipedia is verifiability, not truth... Articles should rely on credible, third-party sources with a reputation for fact-checking and accuracy. For academic subjects, the sources should preferably be peer-reviewed. Reputable sources (academic peer-reviewed textbooks and journals) all support Einstein's relativity. Pfalstad 02:30, 9 October 2006 (UTC)
I'm sorry you lost me there. You defined only what you meant by 'sources'. Therefore: the "reputable", "all", "support", without definitions, shall equal the "time", "time dilation", "wormhole", "fabric of space" within a theory. Now I get where your confusion about relativity comes from: you believe to your beliefs. Get this: if something is published in form of a scientific paper, and subsequently disputed (like the H-K "experiment") in form of another scientific paper, this does not mean that you can disregard the dispute and lead the public (in the above-outlined sneaky ways) to believe that your grandpa's experiment had ANY relevance to proving a theory that (still) cannot account for 98% of the mass of the Universe. About you Einstein trolls being sneaky (concerning your concluding sentence): NO reputable source supports Einstein's THEORY of relativity. Hence, even though you purposely omitted the word 'theory' in your final sentence, Einstein's wannabe laws remain – a theory. Do read more than what your beliefs permit you to, please.
But relevant to this article is that the atomic clocks in the satellites are corrected in order to run in sync with their master clocks on earth. They are set to run slower relative the master before liftoff, and this correction for the speed and gravity in orbit is exactly what is calculated from SR/GR theory. (And you are aware that in physics we do only have theories: ther are no natural laws at all, and there is no point where a theory gets "promoted" to be a law. Everything we have is just and only theories.)Mossig 20:19, 9 October 2006 (UTC)
Of course not. Hence the "Newton's laws", "Kepler's law's", "Bernoulli's law", etc., v. "Einstein's theory", "Planck's theory" to name a few. Concerning your claim that the SV atomic clocks are corrected for GR/SR "corrections": that's a fact that we all know of. The issue here is whether those "corrections" are necessary or not. Leick says not because they simply - cancel out. So PRIOR TO 1990-ies THERE WERE NO SUCH CORRECTIONS APPLIED IN GPS, YET GPS WORKED FINE: "...Changes in the designated Master Control Station (MCS), which maintains the GPS master clock, are not corrected for altitude changes, which cause gravitational relativity shifts. These relativity effects are discussed." (http://adsabs.harvard.edu/abs/1990ion..proc..138D) Now, since the rival side is unable to produce any meaningful scientific citations except for one (the Ashby's) and since I provided at least two opposite scientific views, will some administrator please unlock the GPS and GH-K pages? Thanks in the name of truth, logics AND verifiable scientific facts.
Do you have any references for the claim that the SR/GR-corrections were not done from the start? It directly contradicts the common understanding of the development of the system. (Your statement that there is a differenc between the theories that we call "laws" and "theories" only goes to show that you need to study some scientific theory and history.) And as I have remarked elsewhere: the cite you make does not support your position. Mossig 11:25, 10 October 2006 (UTC)
The reference is above; I hope you are not illiterate. It is YOU who should learn the difference between a theory and a law. It is no one's fault but Mr. Einstein's that he proposed a theory that he hoped applied to everything (therefore -- universal law). The fact that it failed to account for 98% of the mass of the Universe (once humans become able to calculate that estimate as recently done at Caltech), speaks for itself.
I've "known" for some time that the original system had its relativity nay-sayers to the point that the original implementation had a "relativity switch", and it was rapidly shown the switch had to be set to the "Einstein" position. Damned if I can remember where I "learned" this though. A random hit at google-groups provides a hook though: "The idea of correcting for relativity was not obvious to the original GPS designers, mostly military engineers, back in the 1970s. "It was controversial," recalls Ashby, who served as a consultant. "Some people believed you had to account for it; some didn't." So divided were the designers that the first GPS satellite was launched without the frequency offset but had a switch to turn on the offset just in case. It quickly became apparent that the switch had to be on, Ashby says." Hearsay, from a 2004 September Scientific American. While not the place where I "learned" this (I've 'known' since the early 1990's), and sadly from a magazine far past it's prime, but this reference is certainly far superior to the baseless assertions we are obtaining at a glacial pace from Mr. User:Uknewthat. mdf 20:45, 10 October 2006 (UTC)
Did someone say google+Ashby? Oh boy... Add a "relativity switch" to a "time dilation" and a "Warped space" to cook a nice "working hypothesis" that works sometimes but then does not work at some other times, as well. But hey, we all "know" that Mr. Einstein already patented (pun intended) Time. Too bad the Nobel committee members didn't buy it. Or, in words of an Einstein groupie: CITE an exact reference, please! Also, would someone FINALLY define Time, in absolute terms please? What? I didn't think so...Uknewthat 16:50, 11 October 2006 (UTC)
For the sake of sanity: The Nobel commitee is no judges of what in general is science or not. They only give a price to the foremost discovery each year. One basic restriction they are working with is that at most three living people may share a price: thus probably they could not give one for the discovery of SR, as there were more than three people directy involved in the formulation of that theory. Einstein had three more ideas, all worthy of a Nobel price: He got one for the photo-electric effect, could have gotten another one for the brownian motion paper (perhaps?), and the GR was not sufficiently proven during EInsteins lifetime to be eligible. And to go back to the matter at hand: I have still not seen any reference that the SR/GR-corrections were not applied from the start of the GPS system. Please give me a reference, if you have one?
I am sorry, but if you can say that about the Nobel committee (actually: hundreds of most respected scientists who make the jury), then there is nothing for us to discuss. (As for the reference: see the DATE on Mr. Deines's and the subsequent paper that someone cited during this discussion.)
The Deines abstract is dated 1990. What does that prove? From the other Deines abstract, which has been referenced here by myself and others, it's apparent that the "missing relativity effects" he is talking about are the effects of gravity and rotational velocity that arise when a GPS receiver is on the surface of the planet or in an aircraft. The original relativity corrections were calculated for a stationary location at the center of the Earth's gravity. There is nothing anywhere about GPS working for two decades before somebody made any relativity adjustments. Look at the report that mdf posted on the first Phase I GPS satellite at the beginning of the program. Page 180 describes the adjustments made to the clocks in order to compensate for relativity effects. Page 182 lists the achievements of the mission. Achievement #4 is "verification of relativity theory". What is it going to take at this point??? Justin 16:57, 13 October 2006 (UTC)
Iknewthat: Please read my comment above again for comprehension. I have the uttermost respect for the Nobel Price jury (which only consist of 7 people, btw.). And read Justins comment above this one: he has read your reference.Mossig 17:18, 13 October 2006 (UTC)

You keep quoting Leick, Uknewthat, but he never says that relativity corrections are unnecessary. In fact, he says "Relativistic effects are important in GPS surveying, but fortunately can be accurately computed." Putting time dilation in quotes is POV. For sources, see any relativity textbook. Your changes that state "so GR/SR theories are not needed even for high-precision GPS positioning. Hence, GPS cannot be used as a proof of the validity of GR/SR theories" is POV. You have not cited any sources for this. Your source Leick says the exact opposite. That harvard quote is unclear, being just the abstract of a paper, but it seems to be arguing the necessity for relativistic corrections. You provide no source for the claim that "PRIOR TO 1990-ies.... GPS WORKED FINE". This is why the GPS article was locked. Pfalstad 12:30, 10 October 2006 (UTC)

Of course he said it -- implicitly. Read his previously posted sentence on the differencing cancelling out ANY "relativistic corrections" in GPS... To cite a relativity textbook, in making claims opposing relativity -- oh I see that was supposed to be a joke. The rest of your answer deserves no serious comment either. Oh, and do spare us your labelling the things you don't like as "POV". The Harvard quote is very clear to an unbiased reader. Indeed it argues for the corrections, it's just its DATE that does not fit Einstein groupies' claims that the GPS was corrected since day one otherwise it couldn't work. The military had used GPS (to a limited degree, due to smaller global coverage) since 1980-ies.
Saying it implicitly does not cut it. The source must say it explicitly. And he doesn't say the differences cancel out ANY relativistic corrections. He says "most of the relativistic effects cancel or become negligible". Not "all", "most". The Harvard quote does not address any of the points that are in the article. The article does not make any claims about whether GPS corrections were present from the start or not. Pfalstad 03:09, 11 October 2006 (UTC)
Of course it does suffice to make an implicit claim. Otherwise, scientific papers would look something like BASIC or FORTRAN code routines. So, the source does NOT have to say it explicitly. The "most" of a few cm (forget the 11 km relativistic corrections -- they lied to you!) is well below geophysical noise levels, so who cares, let alone knows, whether GPS proves relativity or not? Oh man, just listen to yourself: "The [Harvard] article does not make any claims about whether GPS corrections were present from the start or not" -- look at the complaints the article makes about relativistic corrections MISSING, and then look at the DATE OF THE ARTICLE.Uknewthat 16:53, 11 October 2006 (UTC)
No, not the [Harvard] article, the GPS article that we are working on here. You're not good with the brackets. The Harvard quote has no bearing on it. Pfalstad 17:04, 11 October 2006 (UTC)
Sadly, my local university has no holding for the publication User:Uknewthat cites. However, further searches show that Deines has had more to say about GPS and relativity, and apparently it is wrong. mdf 12:44, 10 October 2006 (UTC)
Do you always post links to corrupt files -- this one can't open in a latest Acrobat Viewer (6.0). Besides, who cares for S.D. Deines later claims? Can you prove that his ABOVE citation was incorrect in dates or claims?
Do not remove indentation. Do sign your comments. And do upgrade your Acrobat to the latest version (hint: it isn't 6.0). And the issue is your unreferenced claim that GPS "does not prove relativity", not when corrections were applied etc. You have offered zero references that support your position. mdf 20:26, 10 October 2006 (UTC)
If the Harvard quote, Leick's book, etc., are all zeroes to you, then try your luck elswhere other than sciences.Uknewthat 16:53, 11 October 2006 (UTC)
It's your creative interpretation of your meager sources that is what we object to. Pfalstad 17:04, 11 October 2006 (UTC)
There is no interpretation; just look at the date of those papers that complain about missing relativistic corrections in GPS. History of scientific endeavour cannot be erased and there is nothing you can do about it, unless your guru’s methodology can help you to travel back in time and change things. Oh, oops...
The Harvard quote cited by Uknewthat is not a peer-reviewed publication. It's a conference proceeding which may be difficult to obtain. However, the author, Steven D. Deines, is listed as an author on a few other citations which mention GPS and relativity. This one is particularly interesting, [11]. Here is the abstract:

The author gives the complete relativity effects as applied to a ground-based receiver's local coordinate time. The current Global Positioning System (GPS) relativity corrections were based on an earth-centered inertial reference frame. The derivation assumed that the receiver obtains inertial GPS coordinate time from the satellites. However, the receiver has been treated tacitly as being stationary in the inertial frame. The problem is that relativity effects in GPS are compensated only for the moving satellites relative to the frame. It is shown that relativity effects for a ground-based receiver include gravity and earth rotation. Airborne GPS receivers have larger effects, and spaceborne GPS receivers have the worst uncompensated relativity effects.

So, Steven Deines, the most recent author that Uknewthat has cited, states explicity that relativity effects are present in the satellites and the receivers. I'm having trouble following the argument here. Is Uknewthat now saying that relativistic corrections are needed now, but they have not always been needed? Please, cite something. Justin 23:23, 10 October 2006 (UTC)
As I told the above admirer of Mr. Einstein's (last) name: look at the DATE of Mr. Deines's paper. But you may keep spitting on Mr. Deines’s credibility as well as on his CITED paper as much as you like, however that is not going to change one bit of WHAT was said there or especially WHEN it was said, nor am I going to give you any credit by fighting your personal animosity towards those who think differently than you and Mr. Einstein-the-patent-clerk. Mr. Deines's paper clearly NEGATES Mr. Ashby's SPECULATION on some "relativity switches" that no one has ever heard of, let alone wrote about in scientific papers.Uknewthat 16:53, 11 October 2006 (UTC)
In his 2002 May article on GPS and relativity in Physics Today (referenced by this very article), Ashby repeats the same story, and cites "Proc. 9th Annual Precise Time and Time Interval Applications and Planning Meeting", L. Rueger et al. (eds.), 1977, J. Buisson, R. L. Easton, T. B. McCaskill, p. 177. I though I had to go and dust off soem NASA technical reports, but the Internet Goddess answered my prayers: "The {T-O) slope gives the frequency offset of +442.5 pp1012 with respect to the PMA clock. Inclusion of the PMA frequency offset of +0.6 PP1012 produces an NTS measured value of +443.1 pp1012. Comparison of this value to the predicted value of the relativistic offset of +445.0 pp1012 gives a difference of -3.1 pp1012. On Day 215, 1977, the NTS-2 PRO-5 output signal was offset {Fig. 21) through the use of a frequency synthesizer {4)." (Emphasis added; further fine-tuning details are presented in the paper) Some speculation. Anyways, I'm now washing my hands of this, since, given his latest statements, I am completely convinced that Uknewthat is beyond reason (cf. USENET kook): as soon as this article is unblocked, the probability he will continue his vandalism is indistinguishable from 1. "The goal of a troll is to crap in your parlor and await comment." He's had enough attention. mdf 18:31, 11 October 2006 (UTC)
Oh, was that supposed to sound inspiring AND funny? So sorry that no one is laughing... Take my advice and stop pretending to be a scientist. Instead, do what is within your abilities and look at the DATE of the GPS papers complaining about LACK of relativistic corrections in GPS. Then ask yourself: oh, mine, how come that GPS worked fine without relativistic corrections for two decades before someone claimed they were a must?
And as mdf showed with this reference, [12], the source Uknewthat cited (and I just quoted) wasn't even correct in his calculations. It's a mess. So tell me why are more reputable sources, such as [13] and [14] being disregarded? Justin 23:35, 10 October 2006 (UTC)
Of course it is not "a mess" as you put it so "politely" -- you have no authority in the subject matter to make such claims, unless you can produce a direct opposing paper as well as a proof of the subsequent withdrawal of the disputed paper by its author(s). I did manage (finally) to open the above file. While being produced at even a later date than Mr. Deines's paper, it also claims a LACK of relativistic corrections in GPS. Therefore, I conclude that this paper only strengthens my point.Uknewthat 16:53, 11 October 2006 (UTC)
I would be in favor of changing the article to state, as the PDF does, that "The effects of relativity... are too small to matter -- less than one centimeter, for users on or near the earth." I think that makes your one good point quite effectively. (Note that the PDF goes on to say that relativity does matter for satellites using GPS data.) The Leick quote just muddies the issue because it's talking about relative positioning. I find your other changes, like putting time dilation in quotes, and saying "GPS cannot be used as proof of GR/SR" to be unacceptable. The only relativity-related claim the GPS article makes (other than linking to the Ashby article) is to talk about the frequency on the satellites being slightly lower than how it is perceived on earth. Do you dispute that? Sources, if so? The PDF file doesn't address it, because it only talks about receivers, not corrections being made on the satellites. The Leick quote doesn't address it either. In fact, in the portion of the Leick quote which you conveniently edited out (p.170), he acknowledges the frequency adjustment. That is proof of GR/SR right there. Pfalstad 17:26, 11 October 2006 (UTC)
I would not object to your first proposition. As for the GPS allegedly being a proof of GR/SR: look, in order for something to stretch or shrink, that "something" first must be definable in absolute terms. But even if you could do it in case of Time (which you cannot), an observed change of time in time (dt/dt) would be producing unity, i.e. would be resulting in nonsense. If the GPS could prove anything, this PERHAPS could refer to SR (of course only as a mathematical alternative), but NOT to GR. People have proposed that simple mechanical drag of atomic orbits (that make the heart of atomic clocks) could affect the atomic clocks' recording of time change (dt), as well. Basically, it all boils down to whether you believe in aether (also referred to by some as vacuum energy, dark energy/matter...) that makes such resistance possible, or in Einstein's PRESUMED empty space. There is no reason whatsoever why both fundamental ASSUMPTIONS would not be represented simultaneously, especially since the jury is still out and the dark energy/matter actually have been observed. Uknewthat 20:26, 18 October 2006 (UTC)
No, You look Uknewthat. From the scientific content of your posts (or lack thereof) it would appear that you have absolutely no grasp on the physical effects of the application of the Theory of relativity. Look at the scientific jargon uses of the words Fact and Theory first, this may go a small way to explaining where you keep going wrong. Also, for something to stretch or shrink it does not need to be defined in absolute terms. Relativity does not assume a single, universal reference point in which everything is defined. If you are travelling, then you are immediately in a different frame of reference to anyone not travelling at exactly the same velocity. The fact that you are not travelling at an appreciable fraction of the speed of light makes no difference to this fact. As you travel faster, the effects of Relativity become more and more pronounced until they become significant to your subjective viewpoint. Time is not an absolute! You should know this already! This Leick bloke does not agree with as has been proved by backtracking on the quotemine you so helpfully provided, and if the makers of the system itself (NASA of all people!) decided during the op/eval period of the GPS network, that GR/SR should be taken into account, that's good enough for me. I'm an undergraduate and I'm running rings around your pseudoscientific claims. (Aether? Yeah, right-_- allow me to introduce you to my good friends Mr Michelson and Mr Morley) I'd hate to see what a good theoretical PhD would do to you. 86.25.2.252 05:18, 4 September 2007 (UTC)

PPS

Pulse per second links here, but the fact that some GPS receivers support PPS is not discussed in the GPS article. Shouldn't this be mentioned and/or linked in the "Precise time reference" section?

GPS recievers are very good PPS sources in the long term sense (the average over many pulses are correct) they are used for syncronising atomic clocks. I second that this should be mentioned in the article. Mossig 08:44, 5 October 2006 (UTC)

GPS and relativity

First of all, User:Uknewthat is a dead-ringer for the dime-a-dozen anti-relativity cranks that infest USENET. If my assessment is correct, any administrator reading this needs to realize that rehabilitation will not work with these individuals. There are kooks that have persisted for literally years in that venue, and now he (and they) have discovered Wikipedia to continue their disruption. Attacking relativity by babbling on about how GPS is not proof of relativity is standard fare for these people. Just check google-groups for a multitude of examples. I strongly suggest that if, post protection, this one persists even one more edit he be shown the door post-haste. At least in my opinion.

That little bit out of the way, we can correct a few misconception here.

The business about the "11km per day" (mentioned variously) just isn't true. This is because fixes are computed from differential measurements of multiple sources subjected to the same perturbations. For a receiver computing stuff every second, the errors might amount 11000/86400 metres, or 12cm.

When thinking about this, it may be useful to consider that for a GPS-like system to function, all that is necessary is that the characteristics of the on-orbit clocks is known to the receiver. For GPS this is done in two ways: (1) the system spec says all the clocks are going to be as identical as they can and (2) what relatively small variances remain are broadcast for the receivers use. While this is just sane system design at work, it is instructive to consider an "insane" system where all the on-orbit clocks are wildly different. Not just trivial things like 38 microseconds a day because of relativistic effects, but seconds or even minutes per day for whatever crazy reason. Despite all of this, it is still possible to construct a perfectly functioning GPS-like system here, but it just becomes a major headache to monitor the ensemble and it would consume alot more bandwidth to downlink the necessary model parameters to the receiver.

So all told, the fact that the on-orbit clocks in the real GPS are tweaked so they appear to run at the same rate as the master control clocks is not because of gravitational redshift and time dilation "need" to be counter-acted, but simple bureaucratic convenience which follows from sensible system design.

The "real" relativistic effects in GPS are in the variations of the observed clock rate as a function of the position in the slightly eccentric orbits. For a single-receiver navigation fix, these corrections do not cancel out like a system-wide skew or time-rate would. The satellites broadcast the associated correction terms (check the signal specification for details; Leick may mention the detail), and they amount to about 50 nanoseconds (about 15 metres), varying on relatively short time-scales (hours). Failure to apply these corrections does in fact produce less accurate positions. Indeed, even the "insane" system would require these adjustments to produce accurate fixes. mdf 21:14, 5 October 2006 (UTC)

I agree in the above. The relativistic drift of 38us per day is not a problem in itself for GPS: it could have been countered in the recievers. But that is not the point that is made when pointing to the GPS system in relation to the theory of relativity: the point is that there is a relativistic time dilation of the clocks in the satellites, which amounts to 38 us/day, and is in perfect correspondence to what SR/GR predicts. This is way GPS is used as an example of a "proof" of SR/GR. (Not that we in a strict sense can ever prove a scientific theory.)
(The error of 11 km/day is what you would get in a very simplistic system using no correction for the relativistic effects. It is a simplified way of showing that 38 us corresponds to a huge error in GPS positioning, as us and ns is not that readily available concepts for the layman.Mossig 23:35, 5 October 2006 (UTC)
I'm sorry, but the 11km/day would not occur even in an "uncorrected" system. Whether or not a "layman" would understand a microsecond, I strongly suggest that articles should not mislead people. mdf 13:37, 6 October 2006 (UTC)
I agree, it shouldnt be mentioned in the article. But I don't think it is either? Mossig 14:35, 6 October 2006 (UTC)
Wikipedia:Requests for comment/Uknewthat is another location for user conduct comments. --Dual Freq 23:21, 5 October 2006 (UTC)
mdf: I thought that Wiki policy does not permit insults like "kook", etc. Anyway, you seem to agree that the "11 km error", allegedly caused by the so-called relativistic effects, is a laughable claim to make indeed. But then I pointed that out first. So why am I a “kook”, and you for example, the Wiki’s most famous bird watcher, are not? (By the way, the rest of your diatribe on “administrative reasons” is mumbo-jumbo plain and simple. I advise you to delete those unnecessary words, and stick to birds). — Preceding unsigned comment added by Uknewthat (talkcontribs) 11:47, 6 October 2006


Yes, if the "11km/day" stuff is being mentioned in the article, then I agree it should be removed. If the 38us per day is being mentioned, no particular problem if mentioned in context: it stands as excellent evidence for SR and GR, whether or not is has any operational effect on GPS-like systems. (The only issue is whether such things need to be said at GPS at all: evidence for and against relativity may be better placed in the relativity articles). That there are other relativistic corrections in the GPS system may not be as well known (I mention them above), but they do exist and they work to within measurement error, and they would be necessary even if the clock-drift (somehow) didn't exist. This is why when you say that "so GR/SR theories are not needed even for high-precision GPS positioning. Hence, GPS cannot be used as a proof of the validity of GR/SR theories.", (understanding that by "proof" you really mean "evidence") and repeat this in various ways several times over a month here and at the HK experiment article, one can't help but get the strong "kook" impression, since it is hard to rationalize this as a simple mis-communication (or mis-reading) on the part of other editors here. Your failure to engage in any substantial discussion is, unfortunately, telling. mdf 13:37, 6 October 2006 (UTC)
I am not going to engage into an insults-war with you. As for the discussion see above my response to Jason. Cheers.
I have no idea who this "Jason" is. Clarification probably isn't useful either, since all of your arguments (to date) appear to be based on misrepresentations of the material you are citing. (An observation that is also consistent with my hypothesis). Let's cut to the chase: are you going to edit as per policy or not? If so, we can ask the lock cut off this article. If not, this matter should be taken to a Higher Power as quickly as possible. Some people probably have work they would like to do here. mdf 19:40, 10 October 2006 (UTC)